在该项目中,你将使用生成式对抗网络(Generative Adversarial Nets)来生成新的人脸图像。
该项目将使用以下数据集:
由于 CelebA 数据集比较复杂,而且这是你第一次使用 GANs。我们想让你先在 MNIST 数据集上测试你的 GANs 模型,以让你更快的评估所建立模型的性能。
如果你在使用 FloydHub, 请将 data_dir 设置为 "/input" 并使用 FloydHub data ID "R5KrjnANiKVhLWAkpXhNBe".
data_dir = './data'
# FloydHub - Use with data ID "R5KrjnANiKVhLWAkpXhNBe"
#data_dir = '/input'
"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
import helper
helper.download_extract('mnist', data_dir)
helper.download_extract('celeba', data_dir)
show_n_images = 25
"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
%matplotlib inline
import os
from glob import glob
from matplotlib import pyplot
mnist_images = helper.get_batch(glob(os.path.join(data_dir, 'mnist/*.jpg'))[:show_n_images], 28, 28, 'L')
pyplot.imshow(helper.images_square_grid(mnist_images, 'L'), cmap='gray')
CelebFaces Attributes Dataset (CelebA) 是一个包含 20 多万张名人图片及相关图片说明的数据集。你将用此数据集生成人脸,不会用不到相关说明。你可以更改 show_n_images 探索此数据集。
show_n_images = 25
"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
mnist_images = helper.get_batch(glob(os.path.join(data_dir, 'img_align_celeba/*.jpg'))[:show_n_images], 28, 28, 'RGB')
pyplot.imshow(helper.images_square_grid(mnist_images, 'RGB'))
由于该项目的重点是建立 GANs 模型,我们将为你预处理数据。
经过数据预处理,MNIST 和 CelebA 数据集的值在 28×28 维度图像的 [-0.5, 0.5] 范围内。CelebA 数据集中的图像裁剪了非脸部的图像部分,然后调整到 28x28 维度。
MNIST 数据集中的图像是单通道的黑白图像,CelebA 数据集中的图像是 三通道的 RGB 彩色图像。
你将通过部署以下函数来建立 GANs 的主要组成部分:
model_inputsdiscriminatorgeneratormodel_lossmodel_opttrain检查你是否使用正确的 TensorFlow 版本,并获取 GPU 型号
"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
from distutils.version import LooseVersion
import warnings
import tensorflow as tf
# Check TensorFlow Version
assert LooseVersion(tf.__version__) >= LooseVersion('1.0'), 'Please use TensorFlow version 1.0 or newer. You are using {}'.format(tf.__version__)
print('TensorFlow Version: {}'.format(tf.__version__))
# Check for a GPU
if not tf.test.gpu_device_name():
warnings.warn('No GPU found. Please use a GPU to train your neural network.')
else:
print('Default GPU Device: {}'.format(tf.test.gpu_device_name()))
部署 model_inputs 函数以创建用于神经网络的 占位符 (TF Placeholders)。请创建以下占位符:
image_width,image_height 和 image_channels 设置为 rank 4。z_dim。返回占位符元组的形状为 (tensor of real input images, tensor of z data, learning rate)。
import problem_unittests as tests
def model_inputs(image_width, image_height, image_channels, z_dim):
"""
Create the model inputs
:param image_width: The input image width
:param image_height: The input image height
:param image_channels: The number of image channels
:param z_dim: The dimension of Z
:return: Tuple of (tensor of real input images, tensor of z data, learning rate)
"""
# TODO: Implement Function
tensor_realinput = tf.placeholder(tf.float32, shape = (None, image_width, image_height, image_channels),
name = 'realinput')
tensor_zdata = tf.placeholder(tf.float32, shape = (None, z_dim), name = 'zdata')
tensor_lr = tf.placeholder(tf.float32, shape = None, name = 'lr')
return(tensor_realinput, tensor_zdata, tensor_lr)
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_model_inputs(model_inputs)
部署 discriminator 函数创建辨别器神经网络以辨别 images。该函数应能够重复使用神经网络中的各种变量。 在 tf.variable_scope 中使用 "discriminator" 的变量空间名来重复使用该函数中的变量。
该函数应返回形如 (tensor output of the discriminator, tensor logits of the discriminator) 的元组。
def discriminator(images, reuse = False):
"""
Create the discriminator network
:param image: Tensor of input image(s)
:param reuse: Boolean if the weights should be reused
:return: Tuple of (tensor output of the discriminator, tensor logits of the discriminator)
"""
# TODO: Implement Function
with tf.variable_scope('discriminator', reuse = reuse):
layer_input = images
layer_hidden_1_input = tf.layers.conv2d(layer_input, 64, 5, strides = 2, padding = 'same')
layer_hidden_1_output = tf.maximum(0.2 * layer_hidden_1_input, layer_hidden_1_input)
layer_hidden_2_input = tf.layers.batch_normalization(tf.layers.conv2d(layer_hidden_1_output,
128, 5, strides = 2, padding = 'same'))
layer_hidden_2_output = tf.maximum(0.2 * layer_hidden_2_input, layer_hidden_2_input)
layer_hidden_3_input = tf.layers.batch_normalization(tf.layers.conv2d(layer_hidden_2_output,
256, 5, strides = 2, padding = 'same'))
layer_hidden_3_output = tf.maximum(0.2 * layer_hidden_3_input, layer_hidden_3_input)
layer_output_input = tf.layers.dense(tf.reshape(layer_hidden_3_output, (-1, 4 * 4 * 256)), 1)
layer_output_output = tf.sigmoid(layer_output_input)
return(layer_output_output, layer_output_input)
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_discriminator(discriminator, tf)
部署 generator 函数以使用 z 生成图像。该函数应能够重复使用神经网络中的各种变量。
在 tf.variable_scope 中使用 "generator" 的变量空间名来重复使用该函数中的变量。
该函数应返回所生成的 28 x 28 x out_channel_dim 维度图像。
def generator(z, out_channel_dim, is_train = True):
"""
Create the generator network
:param z: Input z
:param out_channel_dim: The number of channels in the output image
:param is_train: Boolean if generator is being used for training
:return: The tensor output of the generator
"""
# TODO: Implement Function
flag = not is_train
with tf.variable_scope('generator', reuse = flag):
layer_input = z
layer_hidden_1_input = tf.layers.batch_normalization(tf.reshape(tf.layers.dense(layer_input, 2 * 2 * 512), (-1, 2, 2, 512)),
training = is_train)
layer_hidden_1_output = tf.maximum(0.2 * layer_hidden_1_input, layer_hidden_1_input)
layer_hidden_2_input = tf.layers.batch_normalization(tf.layers.conv2d_transpose(layer_hidden_1_output,
256, 5, strides = 2, padding = 'valid'), training = is_train)
layer_hidden_2_output = tf.maximum(0.2 * layer_hidden_2_input, layer_hidden_2_input)
layer_hidden_3_input = tf.layers.batch_normalization(tf.layers.conv2d_transpose(layer_hidden_2_output,
128, 5, strides = 2, padding = 'same'), training = is_train)
layer_hidden_3_output = tf.maximum(0.2 * layer_hidden_3_input, layer_hidden_3_input)
layer_output_input = tf.layers.conv2d_transpose(layer_hidden_3_output, out_channel_dim, 5, strides = 2, padding = 'same')
layer_output_output = tf.tanh(layer_output_input)
return(layer_output_output)
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_generator(generator, tf)
部署 model_loss 函数训练并计算 GANs 的损失。该函数应返回形如 (discriminator loss, generator loss) 的元组。
使用你已实现的函数:
discriminator(images, reuse=False)generator(z, out_channel_dim, is_train=True)def model_loss(input_real, input_z, out_channel_dim):
"""
Get the loss for the discriminator and generator
:param input_real: Images from the real dataset
:param input_z: Z input
:param out_channel_dim: The number of channels in the output image
:return: A tuple of (discriminator loss, generator loss)
"""
# TODO: Implement Function
gen_model = generator(input_z, out_channel_dim)
dis_model_real, dis_logits_real = discriminator(input_real)
dis_model_fake, dis_logits_fake = discriminator(gen_model, reuse = True)
dis_loss_real = tf.reduce_mean(tf.nn.sigmoid_cross_entropy_with_logits(logits = dis_logits_real,
labels = tf.ones_like(dis_model_real)))
dis_loss_fake = tf.reduce_mean(tf.nn.sigmoid_cross_entropy_with_logits(logits = dis_logits_fake,
labels = tf.zeros_like(dis_model_fake)))
gen_loss = tf.reduce_mean(tf.nn.sigmoid_cross_entropy_with_logits(logits = dis_logits_fake,
labels = tf.ones_like(dis_model_fake)))
dis_loss = dis_loss_real + dis_loss_fake
return(dis_loss, gen_loss)
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_model_loss(model_loss)
部署 model_opt 函数实现对 GANs 的优化。使用 tf.trainable_variables 获取可训练的所有变量。通过变量空间名 discriminator 和 generator 来过滤变量。该函数应返回形如 (discriminator training operation, generator training operation) 的元组。
def model_opt(d_loss, g_loss, learning_rate, beta1):
"""
Get optimization operations
:param d_loss: Discriminator loss Tensor
:param g_loss: Generator loss Tensor
:param learning_rate: Learning Rate Placeholder
:param beta1: The exponential decay rate for the 1st moment in the optimizer
:return: A tuple of (discriminator training operation, generator training operation)
"""
# TODO: Implement Function
dis_vars = [ x for x in tf.trainable_variables() if x.name.startswith('discriminator') ]
gen_vars = [ x for x in tf.trainable_variables() if x.name.startswith('generator') ]
dis_opts = [ x for x in tf.get_collection(tf.GraphKeys.UPDATE_OPS) if x.name.startswith('discriminator') ]
gen_opts = [ x for x in tf.get_collection(tf.GraphKeys.UPDATE_OPS) if x.name.startswith('generator') ]
with tf.control_dependencies(dis_opts):
dis_opt = tf.train.AdamOptimizer(
learning_rate, beta1).minimize(d_loss, var_list = dis_vars)
with tf.control_dependencies(gen_opts):
gen_opt = tf.train.AdamOptimizer(
learning_rate, beta1).minimize(g_loss, var_list = gen_vars)
return(dis_opt, gen_opt)
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_model_opt(model_opt, tf)
"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
import numpy as np
def show_generator_output(sess, n_images, input_z, out_channel_dim, image_mode):
"""
Show example output for the generator
:param sess: TensorFlow session
:param n_images: Number of Images to display
:param input_z: Input Z Tensor
:param out_channel_dim: The number of channels in the output image
:param image_mode: The mode to use for images ("RGB" or "L")
"""
cmap = None if image_mode == 'RGB' else 'gray'
z_dim = input_z.get_shape().as_list()[-1]
example_z = np.random.uniform(-1, 1, size=[n_images, z_dim])
samples = sess.run(
generator(input_z, out_channel_dim, False),
feed_dict={input_z: example_z})
images_grid = helper.images_square_grid(samples, image_mode)
pyplot.imshow(images_grid, cmap=cmap)
pyplot.show()
部署 train 函数以建立并训练 GANs 模型。记得使用以下你已完成的函数:
model_inputs(image_width, image_height, image_channels, z_dim)model_loss(input_real, input_z, out_channel_dim)model_opt(d_loss, g_loss, learning_rate, beta1)使用 show_generator_output 函数显示 generator 在训练过程中的输出。
注意:在每个批次 (batch) 中运行 show_generator_output 函数会显著增加训练时间与该 notebook 的体积。推荐每 100 批次输出一次 generator 的输出。
def train(epoch_count, batch_size, z_dim, learning_rate, beta1, get_batches, data_shape, data_image_mode):
"""
Train the GAN
:param epoch_count: Number of epochs
:param batch_size: Batch Size
:param z_dim: Z dimension
:param learning_rate: Learning Rate
:param beta1: The exponential decay rate for the 1st moment in the optimizer
:param get_batches: Function to get batches
:param data_shape: Shape of the data
:param data_image_mode: The image mode to use for images ("RGB" or "L")
"""
# TODO: Build Model
session, width, height, channel = data_shape
realinput, zinput, lr = model_inputs(width, height, channel, z_dim)
dis_loss, gen_loss = model_loss(realinput, zinput, channel)
dis_opt, gen_opt = model_opt(dis_loss, gen_loss, learning_rate, beta1)
count = 0
with tf.Session() as sess:
sess.run(tf.global_variables_initializer())
for epoch_i in range(epoch_count):
for batch_images in get_batches(batch_size):
# TODO: Train Model
count += 1
batch_images *= 2
zinput_val = np.random.uniform(-1, 1, (batch_size, z_dim))
session = sess.run(dis_opt, feed_dict={realinput: batch_images,
zinput: zinput_val,
lr: learning_rate})
session = sess.run(gen_opt, feed_dict={zinput: zinput_val,
lr: learning_rate})
if count % 100 == 0:
train_loss_dis = dis_loss.eval({zinput: zinput_val, realinput: batch_images})
train_loss_gen = gen_loss.eval({zinput: zinput_val})
print("Epoch {}, DisLoss: {:.4f}, GenLoss: {:.4f}".format(epoch_i,
train_loss_dis, train_loss_gen))
show_generator_output(sess, 25, zinput, channel, data_image_mode)
在 MNIST 上测试你的 GANs 模型。经过 2 次迭代,GANs 应该能够生成类似手写数字的图像。确保生成器 (generator) 低于辨别器 (discriminator) 的损失,或接近 0。
batch_size = 64
z_dim = 100
learning_rate = 0.001
beta1 = 0.5
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
epochs = 2
mnist_dataset = helper.Dataset('mnist', glob(os.path.join(data_dir, 'mnist/*.jpg')))
with tf.Graph().as_default():
train(epochs, batch_size, z_dim, learning_rate, beta1, mnist_dataset.get_batches,
mnist_dataset.shape, mnist_dataset.image_mode)
在 CelebA 上运行你的 GANs 模型。在一般的GPU上运行每次迭代大约需要 20 分钟。你可以运行整个迭代,或者当 GANs 开始产生真实人脸图像时停止它。
batch_size = 32
z_dim = 100
learning_rate = 0.001
beta1 = 0.5
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
epochs = 1
celeba_dataset = helper.Dataset('celeba', glob(os.path.join(data_dir, 'img_align_celeba/*.jpg')))
with tf.Graph().as_default():
train(epochs, batch_size, z_dim, learning_rate, beta1, celeba_dataset.get_batches,
celeba_dataset.shape, celeba_dataset.image_mode)
提交本项目前,确保运行所有 cells 后保存该文件。
保存该文件为 "dlnd_face_generation.ipynb", 并另存为 HTML 格式 "File" -> "Download as"。提交项目时请附带 "helper.py" 和 "problem_unittests.py" 文件。